HazyResearch flash-fft-conv issues

HazyResearch / flash-fft-conv

FlashFFTConv: Efficient Convolutions for Long Sequences with Tensor Cores

Apache License 2.0

260 stars 25 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Error when installed with setup.py

#28 Kyuu9TroppoFumare closed 2 weeks ago
2
CUDA memory illegal access with large batch size

#27 francescocarzaniga opened 2 months ago
2
Does importing the library fix the torch.initial_seed() ?

#26 dlfqhsdugod1106 opened 3 months ago
0
FlashFFTConv Max_seq_len

#25 Moriarty0923 opened 3 months ago
0
The Backwards implementaion

#24 HuuYuLong opened 4 months ago
2
FlashFFTConv can be definitely be implemented on Mamba, right?

#23 5thGenDev opened 6 months ago
2
Can FlashFFTConv be used for Conv2d on PyTorch?

#22 5thGenDev opened 6 months ago
4
Remove dead code that prevents easy installation in Dockerfile without access to CUDA runtime

#21 dwromero opened 6 months ago
0
[bug] CUDA Runtime Error when implicit padding is required

#20 dwromero opened 7 months ago
2
ERROR: CUDA RT call "cudaFuncSetAttribute(&monarch_conv_cuda_32_32_32_kernel<32, 8, 32768, 2, 16, false, 2, 8, 8>, cudaFuncAttributeMaxDynamicSharedMemorySize, 135168)" in line 969 of file .../csrc/flashfftconv/monarch_cuda/monarch_cuda_interface_fwd_bf16.cu

#19 dwromero opened 7 months ago
8
FlashDepthWiseConv1d is not handling padding properly?

#18 dwromero opened 7 months ago
2
Adding python wheels for the package and for the kernels

#17 michaelfeil opened 7 months ago
4
Training Hyena-based Models with FlashFFTConv + Safari

#16 guyjacob opened 7 months ago
2
CUDA Runtime Error at: monarch_cuda/monarch_cuda_interface_fwd.cu:665

#15 CHDev93 closed 7 months ago
1
How does this compare with cuFFT?

#14 cherrybajan opened 8 months ago
1
Can we install FlashFFTConv by pip?

#13 Hambaobao closed 8 months ago
4
Reason for doubling length in both LongConv and FlashFFTConv implementations

#12 nichoffs opened 8 months ago
1
Error during exporting the model

#11 Songweiping opened 8 months ago
0
Implementation of Monarch Mixer with flashfftconv

#10 b-albar closed 9 months ago
2
inference or training

#9 kashif closed 9 months ago
2
Fix typo in readme

#8 radarFudan closed 9 months ago
0
Plans for a Triton implementation?

#7 arnavdantuluri opened 9 months ago
2
monarch_cuda_interface_fwd_bf16.cu failed with invalid argument (1).

#6 jeohalves opened 9 months ago
1
Enable builds without direct torch.cuda availability and support sm89 / sm90.

#5 wanderingai opened 9 months ago
2
Conv1d fp32

#4 Kumbong closed 9 months ago
0
Is FlashDepthWiseConv1d incompatible with torch.autocast?

#3 manuel-tran closed 9 months ago
3
Can input channel be different from output channel？

#2 jimmylihui closed 9 months ago
11
Where is the monarch_cuda file

#1 jimmylihui closed 9 months ago
5